A recently report by Transparency Market Research indicated that global instant coffee earned US$ 28.12 bil. in 2016. Advantages such as relatively low price, easy and quick to prepare reinforce growth estimates at a rate of 4.8% per annum. Electronic nose is an instrument that through an array of chemical sensors identifies and discriminates the sample vapors without destroying it. Six commercial instant coffee, with six batches at least (totaling 46 samples) from the same industry, were analyzed in triplicate by a Portable Electronic Nose (PEN2) with 7 MOS sensors. Static headspace sampling was made with 2.8g of instant coffee in appropriate vials. Operation conditions consisted in a constant flow rate injection of 400mL.min-1, data acquisition time of 60s with intervals of 1s and final purging of 120s between the samples. Before the multivariate analysis, the Savitzky-Golay algorithm (4th order polynomial and 25 points window) was used to obtain the first derivative of the signal. Common Dimensions (ComDim) is a multi-block non-supervised multivariate analysis initially proposed to evaluate sensory data. To apply the ComDim to e-nose data, a table for each sensor with the averaged 1st derivative signal for each sample was used as input. Pareto chart and salience table (weights for each sensor in each common dimension) indicate that 3 common dimensions (CD) were sufficient to represent 98.3% from data variance. Saliences table indicate a higher correlation in CD1 between sensors 1, 3, 5, 6 and 8. CD2 is explained by sensors 7 and 9. The scores from first four CD (99.9% variance) were used as input to construct a linear discriminant analysis (LDA) classifier. The model reached a sensitivity and specificity of 100% in the leave-one-out cross-validation. Thus, the proposed approach was able to correct identify the aromatic pattern of different instant coffees.